论文部分内容阅读
介绍了在V-REP仿真平台下使用Kinect视觉传感器获取飞机舱门手柄位姿的方法。该方法根据Kinect传感器的RGB彩色相机获取舱门手柄上的两个环形彩带图像并进行Blob Detection处理,得到彩环在Kinect图像平面中的矩形框中心位置,然后根据Depth深度相机检测出彩环的深度信息,由位置和深度信息即可确定矩形框中心点在世界坐标系中的三维坐标。通过圆柱面拟合方法得到圆柱面参数,结合两矩形框中心点进而求出圆柱手柄的重心及姿态信息,实现了舱门机构耐久性试验中,工业机器人对舱门手柄的视觉定位。
The method of using Kinect vision sensor to obtain the attitude of aircraft door handle under V-REP simulation platform is introduced. According to the Kinect sensor’s RGB color camera, the method obtains two annular ribbon images on the handle of the door and carries out Blob Detection processing to obtain the center of the rectangle of the color ring in the Kinect image plane, and then detects the depth of the color ring according to Depth depth camera Information, from the position and depth information to determine the three-dimensional coordinates of the rectangular box center point in the world coordinate system. Cylindrical surface parameters are obtained by fitting cylindrical surface method, and the center of gravity of the two rectangular frames are combined to obtain the center of gravity and attitude information of the cylindrical handle. The visual positioning of the handle of the door by the industrial robot during the durability test of the door mechanism is achieved.